Smoke movement

Smoke movement is the transport of combustion products through a building or other enclosed space. It includes the rising fire plume, horizontal ceiling flow, accumulation in an upper layer and movement through openings to adjacent spaces. Temperature, pressure, geometry and ventilation determine the paths and rates of that transport.

Smoke does not simply fill every space uniformly. Entrainment adds air to the plume, layers form and mix, and openings create jets or counter-current flows. Building shafts, ducts and leakage paths can carry smoke far from the fire enclosure.

Understanding these paths is necessary for detection, evacuation and smoke-control design. The concentration and arrival time at a particular location depend on the fire source and the route travelled. A single room-average description cannot represent every location or stage.

Plume rise and air entrainment

Buoyancy makes hot combustion products rise above the fire. As the plume rises, it draws in surrounding air and carries smoke with it. IFEG explains that the entrained air mass usually greatly exceeds the mass of fuel burned, which is why plume entrainment is central to smoke-production estimates.

Entrainment increases the volume and mass flow transported upwards while changing temperature and concentration. It links fire behaviour to the amount of smoky gas reaching a ceiling or exhaust system. The heat-release rate alone does not describe the entire smoke-transport process.

The fire's position changes access to entrained air. IFEG notes that walls and corners restrict entrainment compared with a fire away from those boundaries. Plume models developed for unrestricted fires therefore require an appropriate treatment for such locations.

Plume correlations also have defined regions and geometry. IFEG distinguishes weak and strong plumes and different calculation methods above a fire. These methods provide estimates for their stated conditions, not universal descriptions of every flame and smoke source.

When the plume reaches a ceiling, its vertical motion turns into a ceiling jet. Hot gases spread laterally and can activate detectors or sprinklers along that path. The ceiling jet is thus an early transport route before smoke becomes distributed through a larger volume.

Layers, mixing and changing concentration

Smoke collecting beneath a ceiling forms an upper layer above relatively cooler air. The height of its interface changes with time, fire size and exhaust. A layer may deepen as more smoke arrives or be maintained at a higher level when removal balances part of the supply.

Layer height is not determined by ceiling height alone. Larger production can supply more smoke, while smoke exhaust removes part of the flow. The enclosure volume and geometry determine how the remaining products accumulate.

The boundary between smoke and clear air is not always perfectly sharp. Mixing at openings, plume motion and mechanical flows can entrain additional air and redistribute particles. This means a simple two-zone description is an approximation whose suitability depends on the scenario.

Sprinkler sprays can cool hot gases and stir the layer. NIST describes the resulting reduction in buoyancy and a more uniform concentration within a compartment in its WTC smoke-management review. This is an interaction, not an assumption that sprinklers remove all smoke.

Smoke properties can also vary along the path as hot products mix with ambient air. A source concentration cannot simply be assigned to a remote room without considering dilution and transport. The assessment should follow the physical route from origin to exposed location.

Pressure-driven flow through openings

Horizontal movement through a doorway or other opening is governed mainly by pressure difference. Hot gas, wind, fans and stack pressures can all contribute to that difference. The resulting direction and magnitude can vary with height above the floor.

An opening can carry opposing flows at different heights. Hot smoke may leave through the upper part while cooler air enters below. IFEG identifies entrainment into the door jet and mixing caused by counter-current flow as processes that must be represented.

Fire-induced expansion and buoyancy contribute to the pressure field. Fire-induced pressure therefore affects both flow within the fire enclosure and transfer to neighbouring spaces. It cannot be evaluated independently of the available openings and their positions.

A Smoke barrier restricts these paths by providing a continuous surface and protected closures. Walls, floors, ceilings and dampers can form compartments that limit smoke movement. Their effect depends on leakage and the condition of doors and other openings.

A closed barrier changes the path rather than eliminating pressure. Pressure differences can drive smoke through remaining cracks or open transfer routes. The complete leakage network is therefore more informative than the nominal presence of a wall or damper.

Vertical paths and stack effect

Vertical smoke movement can occur through horizontal ceiling vents and vertical spaces such as lift shafts, stairwells and service ducts. These connections can link several floors and provide routes far beyond the enclosure of origin.

Stack effect arises when indoor and outdoor temperature differences create buoyancy-related pressure differences in a tall building. The resulting flows depend on building height, openings and leakage. Smoke can be carried with those flows through connected spaces.

Stack effect is not the same as the plume's initial buoyancy. The plume is driven by hot combustion products above the fire, whereas stack effect concerns the building-scale indoor-outdoor temperature difference. Both can act on the same smoke cloud at different stages.

NIST identifies wind as another influence on horizontal and vertical building flows. Open or broken windows can change the leakage network and strengthen wind effects. A building that is relatively closed in normal conditions can therefore experience different smoke movement during a fire.

The Ideal gas law provides the relationship used to connect gas temperature and density in smoke-flow methods. It supports the physical description of buoyancy but does not by itself predict a building's smoke path. Openings, pressure boundaries and fire inputs remain necessary.

Building services and smoke control

Heating, ventilation and air-conditioning systems can transport smoke through ducts and supplied air. Recirculation can distribute products between compartments if the system continues operating in an unsuitable mode. Fire-mode shutdown or redirection changes this transport path.

Fans used for normal ventilation are not automatically smoke-control equipment. Their fire-mode operation, temperature exposure, control and interaction with dampers must match the intended function. Shutdown can also leave an open duct through which smoke continues to move.

Smoke control introduces deliberate airflow, extraction or pressure differences to limit and direct movement. Natural and mechanical venting can remove upper-layer gases while replacement air enters below. The resulting path depends on both the exhaust and its replacement-air arrangement.

Pressurisation aims to establish flow from a protected zone towards a less protected zone. Open doors and pressure relief alter that flow, so excessive pressure can interfere with door operation. The design must balance smoke restriction with escape-route use.

Sprinklers and smoke-control systems interact as well. Cooling can reduce buoyancy, while spray can delay heat-operated vents or later sprinklers. Cooper calls such delayed activation skipping and explains that droplet transport in the ceiling jet contributes to it.

Modelling movement through a building

A Network model represents connected spaces and flow paths as a network. It is suited to analysing pressure-driven movement through multiple openings rather than resolving every detail of a plume. Its result depends on the zones, connections and boundary pressures supplied to it.

Hand calculations can estimate selected quantities such as smoke production, layer height or opening flow. IFEG presents these methods with assumptions about plume geometry and layer development. Their equations should not be transferred beyond the stated fire and enclosure conditions.

Computer models can combine more of the changing fire and building conditions. They still require appropriate source data, geometry and ventilation inputs. Using a more detailed model does not remove uncertainty in those inputs or in the physical approximations.

Movement beyond the fire enclosure is especially difficult to estimate with simple methods. IFEG states that hand calculations for other enclosures are rarely used and have greater uncertainty. Multiple flow paths and changing pressures make remote predictions more demanding.

A useful model follows a defined scenario through time. Ignition and plume rise establish the source, ceiling transport and openings distribute it, and building systems modify the flow. The analysis should preserve that sequence rather than assigning one static smoke distribution to the entire building.

Exposure and engineering interpretation

For evacuation, the relevant questions include when smoke reaches a route, how deep the layer becomes and whether the route remains usable. A source fire may be remote from the exposed occupants, so travel through the building must be included. Detection timing and egress timing depend on that transport.

For firefighting, smoke movement affects visibility and access to the seat of fire. Cooper identifies this as one objective of designed venting. Removing upper-layer smoke may improve access, but only if the resulting airflow does not create unacceptable conditions elsewhere.

For property protection, smoke and hot gases can affect spaces beyond the flame boundary. The route and duration of exposure determine which compartments are affected. A smoke boundary can reduce that spread without necessarily providing the same fire-resistance function as a fire wall.

Model outputs should be interpreted by location and time. A predicted layer height in one zone does not describe every connected room, and a pressure difference across one opening does not describe the entire network. Reporting the scenario and its assumptions prevents overgeneralisation.

Smoke movement is consequently not a single velocity or direction that applies throughout a fire. It is a changing transport process shaped by the fire and the building. Engineering measures work by modifying that process under stated conditions, not by assuming smoke will always rise or remain above occupants.

Variability and connected flow paths

A building can contain several simultaneous pressure-driven paths. Smoke may leave above a doorway while replacement air enters below, and another path may carry products into a shaft. Focusing on only one opening can miss the route that determines exposure on another floor.

Changes to the opening network alter that result. Closing a door restricts one leakage path, while a broken window can introduce wind pressure and a new connection to outdoors. Smoke movement should therefore be assessed for the conditions expected during the fire, not only the normal building layout.

Vertical shafts can connect several enclosed levels. NUREG-1805's description of smoke-network models represents stairwells and other shafts as vertical series of spaces. This allows pressure differences and leakage flows to be followed between floors rather than treating the building as one open volume.

Network calculations use interior and exterior temperatures, the flow network and ventilation flows as inputs. They can provide pressures and airflows throughout the represented building. Their output remains limited by the accuracy of leakage descriptions and the suitability of the network approximation.

Mechanical smoke control modifies those network flows deliberately. Exhaust from one zone and supply to another can change pressure direction across their shared boundary. The effectiveness of that arrangement depends on all parallel leakage paths, not only the duct containing the fan.

Natural smoke venting likewise changes the network. Upper openings release hot gases and lower openings provide replacement air, with buoyancy helping to drive the flow. The direction at each opening depends on its position and the pressure conditions at that time.

Smoke arrival at a detector or occupant location is a time-dependent outcome. It follows plume development, transport through connected spaces and any mixing along the route. This sequence cannot be reduced to a single building-wide delay without losing the differences between locations.

For engineering interpretation, each flow arrow should have a physical cause and a defined path. Temperature-driven buoyancy, fan pressure, wind and stack effect can reinforce or oppose one another. Identifying those contributions prevents a ventilation diagram from being mistaken for a guaranteed smoke pattern.

A smoke-control design changes selected terms in this transport problem. A new exhaust path can remove gas and change pressures, while a closed damper can redirect flow towards another leakage route. The consequence should be followed through the whole connected network to identify where the smoke goes next.

This network perspective also explains why a remote space can be affected without flames leaving the fire room. Combustion products can travel through shared air paths while the fire remains localised. Evaluating only flame spread would therefore miss an important form of exposure.

See also

• Convection
• Fire damper
• Fire door
• Positive pressure ventilation

References

• Australian Building Codes Board. (2020). Fire safety verification method data sheets (Version 1.1). Author.
• Australian Building Codes Board, National Research Council of Canada, International Code Council, & Department of Building and Housing New Zealand. (2005). International fire engineering guidelines (2005 ed.). Australian Building Codes Board.
• Cooper, L. Y. (2002). Smoke and heat venting. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed.). National Fire Protection Association.
• Evans, D. D., Peacock, R. D., Kuligowski, E. D., Dols, W. S., & Grosshandler, W. L. (2005). Active fire protection systems (NIST NCSTAR 1-4). National Institute of Standards and Technology.
• U.S. Nuclear Regulatory Commission. (2005). Fire dynamics tools: Quantitative fire hazard analysis methods for the U.S. nuclear regulatory commission fire protection inspection program (NUREG-1805). Author.